U.S. patent number 4,442,544 [Application Number 06/281,816] was granted by the patent office on 1984-04-10 for adaptive thresholder.
This patent grant is currently assigned to Xerox Corporation. Invention is credited to John F. Moreland, James C. Stoffel.
United States Patent |
4,442,544 |
Moreland , et al. |
April 10, 1984 |
Adaptive thresholder
Abstract
An adaptive thresholder for processing image pixels. A portion
of the image pixel line or lines is examined and the maximum and
minimum pixels in the line portion determined. The maximum and
minimum pixels are differenced, and the resulting difference signal
compared to a constant representing the minimum desired change in
background level. Where the difference signal is larger, a new
threshold is thereafter used. To determine the new threshold level,
the difference signal is multiplied by a factor representing the
amount of change in threshold for a given event and the result
subtracted from the previously determined maximum pixel. The
subtracted output is brought within predetermined maximum and
minimum threshold constraints to provide the new threshold.
Inventors: |
Moreland; John F. (Fairport,
NY), Stoffel; James C. (Rochester, NY) |
Assignee: |
Xerox Corporation (Stamford,
CT)
|
Family
ID: |
23078915 |
Appl.
No.: |
06/281,816 |
Filed: |
July 9, 1981 |
Current U.S.
Class: |
382/273 |
Current CPC
Class: |
H04N
1/403 (20130101) |
Current International
Class: |
H04N
1/403 (20060101); G06K 009/00 () |
Field of
Search: |
;382/53,68 ;250/23R,209
;328/68,78,146,151 |
References Cited
[Referenced By]
U.S. Patent Documents
Other References
Dynamic Video Thresholding, H. Klepp, IBM Technical Disclosure
Bulletin, vol. 22, No. 3, Aug. 1979, pp. 1031-1033..
|
Primary Examiner: Wise; Edward J.
Attorney, Agent or Firm: McMullen; Frederick E.
Claims
We claim:
1. The method of determining the optimum threshold for thresholding
image pixels, comprising the steps of:
(a) buffering a block if image pixels in a stream of image pixels,
at least one of the pixels in said block of pixels comprising the
pixel to be thresholded;
(b) identifying maximum and minimum pixels in said block of
pixels;
(c) differencing said maximum and minimum pixels from one another
to provide a difference signal;
(d) multiplying said difference signal by a predetermined constant
representing a desired change in threshold level to provide a base
threshold; and
(e) differencing said base threshold from said maximum pixel to
provide said optimum threshold.
2. The method according to claim 1 including the steps of:
(a) establishing preset maximum and minimum threshold levels;
and
(b) limiting said optimum threshold to a level no greater than or
less than said preset maximum and minimum threshold levels.
3. The method according to claim 1 or 2 including the steps of:
(a) comparing said difference signal with a second predetermined
constant representing the minimum acceptable rate of change in
image background density to provide a reference signal; and
(b) updating the existing threshold with said optimum threshold in
response to a predetermined difference between said difference
signal and said reference signal to provide a new threshold for
thresholding said one pixel and subsequent pixels.
4. The method according to claim 1 including the step of buffering
at least two lines of image pixels.
5. Apparatus for adaptively thresholding image pixels, comprising,
in combination:
(a) buffer means for storing a plurality of successive ones of said
pixels including the pixel to be thresholded;
(b) means for identifying the maximum pixel in said plurality of
pixels;
(c) means for identifying the minimum pixel in said plurality of
pixels;
(d) first subtractor means for subtracting the output of said
maximum pixel identifying means from said minimum pixel identifying
means;
(e) multiplier means for multiplying the output of said first
subtractor means by a predetermined constant representing a desired
change in threshold level; and
(f) second subtractor means for subtracting the output of said
multiplier means from said maximum pixel identifying means to
provide a threshold for use in thresholding said one pixel.
Description
This invention relates to a method and apparatus for thresholding
image pixels, and more particularly to an improved method and
apparatus for adaptively thresholding image pixels in response to
changes in image content.
In image reading systems employing for example a raster scanner
having one or more linear arrays of Charge Coupled Devices
(commonly referred to as CCD's) for scanning document originals,
the varying background densities of different pages and the
different subimage areas of the same page, especially magazines
with color, can present problems. In addition, the density of the
printing inks and, hence the line copy and halftones produced can
also vary. As a result, it is not always possible to select a
single density level at which to threshold the raster scanner
output which is ideal for all paper and ink conditions. This
problem normally confronts all binary output scanners.
Various techniques have been suggested to overcome or at least
alleviate the problem such as sensing the edge of the document
original and adjusting background level in response to the
background signal obtained, or setting threshold levels as a
percentage of signal peaks detected or by the signal level of
certain ones or combinations of neighboring image areas, etc.
Where for example the document original comprises line copy, i.e. a
typed sheet, variations in input line density may require
variations in threshold. Otherwise, image portions may be missed as
for example where the line density is below the existing threshold
level. In that case, the line image is consigned to background and
therefore lost. Similarly, variations in background density may
require variations in thresholds. Otherwise, areas that should be
identified as background are instead identified as image areas.
Typically this occurs when the background area rises above the
threshold level as may occur when the document original or a
portion thereof has a colored background. Combinations of the
aforementioned, i.e. variations in line density and in background
density, also demonstrate the need for adjusting threshold levels
in accordance with the image content of the document original.
Where the document original is a halftone, threshold variation is
helpful when one is seeking to reproduce the halftone without
rescreening. There variations in printing density and background
will result in grayscale translations of the type discussed if the
density of the halftone dots produced by the image producer such as
a xerographic machine is different from that of the input.
In considering macro vs. micro adaptive thresholding, macro
adaptive thresholding, which is a once per page correction for the
environment, corrects for background variations and line density
variations. However, if the input image has various densities of
lines for example, an optimal single level threshold may not be
sufficient as noted earlier. Furthermore, the simultaneous
enhancement of detectability for small "white" and "black" detail
is not enabled by a one-per-page, macro adaptive, threshold. In
micro adaptive thresholding, the conversion from multivalued to
binary valued picture elements or pixels is a function of regional
properties of the image. Microadaptation may correct for line
density and background as well as facilitate enhanced detection of
MTF degraded detail imagery. Furthermore, images with regions of
various densities of lines may be reproduced with regionally
independent thresholds. However, micro adaptive thresholding logic
is in general a more complex function than the simple threshold
function. Simple thresholding is a point process while micro
adaptive thresholding is a function of the neighborhood around the
pixel being converted to a binary level.
One and two pass processing may also be thought of where a prescan
of an image is the first step in a macro adaptive strategy. One may
then use a pixel graylevel histogram to select the threshold for
the entire page. This results in a two pass scan of the image or a
grayscale buffering of the entire image. Either method is
relatively expensive. Furthermore, average statistics such as from
a graylevel histogram may be very misleading to threshold selection
logic. A one pass macro adaptive threshold function relies in
general on the image margins and perhaps on the first "black"
information to set the threshold value for the remainder of the
page. Micro adaptive threshold logic may utilize large contexts to
define the binary output value of a single pixel. Although it is
possible to develop a two pass micro adaptive algorithm, in general
only one pass is required. The expense is incurred by the need to
buffer and manipulate the context around the pixel to be converted
to a binary value.
The present invention relates to a method of determining the
optimum threshold for thresholding image pixels, comprising the
steps of buffering a block of image pixels in a stream of image
pixels, at least one of the pixels in the block of pixels
comprising the pixel to be thresholded; identifying maximum and
minimum pixels in the block of pixels; differencing the maximum and
minimum pixels from one another to provide a difference signal;
multiplying the difference signal by a predetermined constant
representing a desired change in threshold level to provide a base
threshold; and differencing the base threshold from the maximum
pixel to provide the optimum threshold.
The invention further relates to an apparatus for adaptively
thresholding image pixels, comprising in combination: buffer means
for storing a plurality of successive ones of the pixels including
the pixel to be thresholded; means for identifying the maximum
pixel in the plurality of pixels; means for identifying the minimum
pixel in the plurality of pixels; first subtractor means for
subtracting the output of the maximum pixel identifying means from
said minimum pixel identifying means; multiplier means for
multiplying the output of the first subtractor means by a
predetermined constant representing a desired change in threshold
level; and second subtractor means for subtracting the output of
the multiplier means from the maximum pixel identifying means to
provide a threshold for use in thresholding the one pixel.
In the drawings:
FIG. 1 is a graph plot illustrating typical image line copy
profiles;
FIG. 2 is a graph plot illustrating typical image background
densities;
FIG. 3 is a circuit schematic showing details of the adaptive
thresholder of the present invention; and
FIG. 4 is a graph showing the adaptive threshold maximum and
minimum relationships.
As used herein, the term pixel refers to a picture element which
essentially comprises a signal, the voltage level of which is
representative of image graylevel. By way of example, a voltage
level of zero may be considered as representing black while a
voltage level of 255 millivolts (m.v.) may be considered as
representing white. Voltage levels between zero and 255 m.v.
represent different shades of gray from the darkest gray at 1
millivolt and progressing upwardly to the lightest gray at 254
millivolts.
The image pixels may be derived from any suitable source as for
example a raster input scanner of the type employing one or more
linear arrays, for example CCD's, arranged to scan the image on a
document original line by line. For this purpose, relative movement
as by means of a document transport is established between the
original being scanned and the array or arrays. The image pixels
produced by the array or arrays, which due to the inherent
operating characteristics of arrays of this type are in bursts of
one line each, are typically processed by thresholding or
screening. For thresholding, the image pixels are compared, i.e.
thresholded against a preset signal or threshold level which
effectively converts the analog pixel signals output by the array
or arrays to binary level pixel signals. In the example given
above, a threshold of 128 millivolts may be envisioned. With the
use of a comparator circuit, pixels below 128 millivolts are
classified as black image areas while pixels equal to or above 128
millivolts are classified as white or background image areas.
Screening is similar except that multiple ones of the image pixels
are compared or thresholded against a corresponding number of
selected threshold levels.
Referring now to FIG. 1 of the drawings, there is shown a plot of a
typical image line, identified by the numeral 10, as derived from
an original having a line copy image, i.e. a typed page. As seen
there, the normal image line A-B rises above the nominal threshold
level and hence is identified by the thresholder as image data.
Areas below the threshold level are of course identified as
non-image or background areas. However, abnormal image lines such
as low density lines shown at C-D, may not rise to the threshold
level and when processed are not identified as image areas but
rather are classified erroneously as non-image or background
areas.
In FIG. 2, a plot of typical image background densities is shown.
There it can be seen that an area E-F of relatively high background
density may be erroneously identified as an image area because of
the background density is above the threshold level in use.
The present invention seeks to accommodate for this inherent
inability in threshold systems, and screening systems as well, to
detect or compensate for marginal image and background areas. The
present invention as will appear provides a threshold level which
adapts or changes in accommodatiofn of variations in image and/or
background artifacts.
Referring to FIGS. 3 and 4, the adaptive thresholder 11 of the
present invention is thereshown. Thresholder 11 includes a buffer
array 12 comprised of a plurality of one pixel delay buffers 13-1,
13-2, 13-3, 13-n, for imposing a succession of one pixel delays on
the stream of image pixels input thereto through line 14 from a
suitable source (not shown). Output line 15 couples the output of
the last buffer 13-n in the buffer array 12 to one input of a
comparator or thresholding circuit 20. Pixel input line 14 is
tapped by line 22 which is in turn coupled to one input terminal of
minimum and maximum pixel level detection circuits 24, 26,
respectively. Lines 27, 28, 29, 30 couple the output side of the
several one pixel delay buffers 13-1, 13-2, 13-3, 13-n comprising
buffer array 12 to the input terminals of minimum and maximum pixel
detection circuits 24, 26.
Minimum and maximum pixel level detection circuits 24, 26 comprise
any suitable circuitry capable of identifying the minimum and
maximum pixels in the block of pixels across delay buffer 12.
Circuits 24, 26 for example may comprise a series of comparator
circuits arranged to compare the various signal inputs thereto for
the purpose of identifying the smallest (i.e. the minimum) and
largest (i.e. the maximum) pixels.
Lines 34, 35 couple the output terminals of minimum and maximum
pixel level detector circuits to differencing circuit herein a
subtractor 38. Subtractor 38, which comprises any suitable circuit
for subtracting or differencing two signal inputs, outputs a
difference signal (RANGE) via line 40 to event detection or
comparator circuit 42 and via line 44 to multiplier circuit 46.
Comparator circuit 42, which controls enabling of new threshold
control gate 50, compares the difference signal output by
subtractor 38 with a preset value .theta.1, representing a minimum
slope or rate of change in image background level, input through
line 47. One suitable value for .theta.1 is 30.degree.. The control
signal output of comparator 42, which controls enabling of new
threshold control gate 50, is fed to the enabling input of gate 50
through line 48.
A threshold determining circuit 45 includes a multiplier 46, which
comprises any suitable multiplier circuit, to multiply the
difference signal output of subtractor 38 by a preset weighting
factor input through line 52. Weighting factor .eta.2 is a value,
which may be experimentally determined, representing the amount of
change in threshold level for a given event. A preferred value of
.theta.2 is 40%. The output of multiplier 46 is coupled by line 55
to one input of subtractor 56. A second input of subtractor 56 is
coupled by line 58 to the output of maximum pixel level detection
circuit 26. Subtractor 56 subtracts or differences the maximum
pixel level output of circuit 26 from the maximum background signal
output of multiplier 46 to provide a new threshold.
The new threshold output by subtractor 56 is fed through line 60 to
threshold level limiting circuit 65. Preset high and low threshold
level limiting voltages are input to circuit 65 through lines 66,
67 respectively. Threshold level limiting circuit 65, which
comprises any suitable comparator circuit, serves to compare the
new threshold output by subtractor 56 with the preset high and low
threshold level limits. Where the new threshold is greater than or
less than one of the preset high and low threshold level limits,
the new threshold is reduced or raised as appropriate to the preset
high or low threshold level limits.
The signal output of threshold level limiting circuit 65 is fed
through line 70 to the input of threshold control gate 50. The
output of gate 50 is coupled to a second input terminal of
threshold circuit 20 through line 72.
OPERATION
Lines of image pixels are clocked pixel by pixel through buffer
array 12 by suitable pixel clock signals, it being understood that
with each clock pulse, the pixels advance one step from one pixel
delay buffer to the next. At each pixel step, the instantaneous
pixel levels in buffer 12 are jointly input to minimum and maximum
pixel detector circuits 24, 26 respectively, where the current
minimum and maximum pixels are identified. The minimum and maximum
pixels identified by circuits 24, 26 are input to subtractor 38
where the pixels are differenced to provide a difference (i.e.
RANGE) signal. The difference signal output by subtractor 38 is fed
to event comparator 42 to determine if the new threshold (which is
calculated in any event) will be used, and to multiplier 46 of
threshold determining circuit 45 where the new threshold is
calculated.
To determine if the new threshold is to be used, comparator 42
compares the difference signal with a predetermined constant,
.theta.1 representing a minimum change in background density. Where
the difference signal is above the constant, indicating that the
rate of change in the background image is greater than the
predetermined minimum represented by the constant .theta.1, a
signal from comparator 42 enables threshold control gate 50 to
apply the new threshold output by threshold determining circuit 45
to thresholding circuit 20. Where the rate of change in image
background is less than the predetermined minimum, control gate 50
is not enabled and application of the new threshold to thresholding
circuit 20 is precluded.
Concurrently, a new threshold is determined based upon the content
of delay buffer 12 by threshold determining circuit 45. For this
purpose, the difference signal (RANGE) output by subtractor 38 is
fed to multiplier circuit 46 of circuit 45 where the difference
signal is multiplied by the weighting factor .theta.2 representing
the amount of change in threshold for a given event. The threshold
signal output by multiplier 46 is fed to subtractor 56 where the
threshold signal is subtracted from the maximum pixel signal output
by maximum pixel detector circuit 26 to provide a new threshold.
The new threshold output by subtractor 56 is compared with preset
maximum and minimum threshold limits and reduced or increased as
required to bring the new threshold to a level no greater than or
less than the preset threshold limits by comparator circuit 65. The
new threshold is output to threshold control gate 50 for
application to thresholding circuit 20 should control gate 50 be
enabled as described heretofore.
While a single line (L1) of pixels is illustrated as being
processed herein, multiple lines (L2, . . . LN) may be processed.
In that event additional buffer arrays would be provided with one
of each additional buffer array being associated with each
additional pixel line.
While the invention has been described with reference to the
structure disclosed, it is not confined to the details set forth,
but is intended to cover such modifications or changes as may come
within the scope of the following claims.
* * * * *